This revised competing continuation tests whether a gradual reduction treatment we found effective for smokers not interested in quitting in our prior grant period will also be effective as a treatment for smokers who are trying to quit. Most treatment guidelines, reviews and tobacco control programs recommend abrupt cessation over gradual cessation yet many smokers prefer to quit gradually. Although recent work suggests gradual cessation may be effective, the five experimental studies of gradual vs. abrupt cessation have produced inconsistent results and had methodological problems such as small sample sizes. We hypothesize gradual cessation is more efficacious than abrupt cessation when aided by nicotine replacement. We will randomize 750 smokers to receive gradual cessation counseling (aided by pre-cessation nicotine lozenge) vs. abrupt cessation counseling vs. no counseling prior to the quit date. The first two conditions will receive 60 min of phone counseling prior to the quit date. After the quit date, all three conditions will receive nicotine lozenge for cessation and one 10 min counseling phone call. Our major outcome will be 6-month prolonged abstinence. We also hypothesize that significant adverse events are not greater when using lozenge with gradual cessation than with abrupt cessation. Prior to the trial, pilot work will include focus groups, a small survey, development of treatment manuals, recruitment strategies and mediator measures and a pilot rehearsal of the final protocol. This revision adds procedures to increase and verify treatment fidelity and decreases the number of differences between the two active treatment groups. If we find gradual cessation produces higher quit rates than abrupt cessation, this would suggest cessation services should encourage gradual cessation. If we find abrupt cessation is superior to gradual cessation, this would be the most comprehensive verification of the current practice of recommending abrupt cessation for all smokers. If we find gradual and abrupt cessation produce similar outcomes and if gradual cessation counseling is better than no counseling, this would suggest gradual cessation should be offered to those who request it. This study will be the first large, systematic test of gradual vs. abrupt cessation.